© Neuronpedia 2026
    Privacy & TermsBlogGitHubSlackTwitterContact
    Neuronpedia logo - a computer chip with a rounded viewfinder border around it

    Neuronpedia

    Natural Language
    Autoencoders
    NEW    Assistant AxisNEWCircuit TracerUPDATESteerSAE EvalsExportsAPI Community BlogPrivacy & TermsContact
    1. Home
    3. Gemma-2-27B
    5. 10-GEMMASCOPE-RES-131K
    7. 105581
    Prev
    Next
    INDEX
    Explanations

    nullification, negation, canceling out

    np_acts-logits-general · gemini-2.5-flash-lite

    The neuron detects words that express negation or cancellation (e.g. nullify, invalidate, negate, cancel out).

    oai_token-act-pair · o4-miniTriggered by @jyhe0408
    New Auto-Interp
    Top Features by Cosine Similarity
    Configuration
    google/gemma-scope-27b-pt-res/layer_10/width_131k
    Prompts (Dashboard)
    24,576 prompts, 128 tokens each
    Dataset (Dashboard)
    monology/pile-uncopyrighted
    No Configuration Found
    Embeds
    IFrame
    Link
    Not in Any Lists

    No Comments

    Negative Logits
     to
    -2.23
     have
    -1.45
    嵯
    -1.40
     â
    -1.39
    、“
    -1.38
    ner
    -1.33
     уже
    -1.33
     zwierząt
    -1.32
    すでに
    -1.30
     at
    -1.27
    POSITIVE LOGITS
     Beamten
    1.73
     possibility
    1.71
     übri
    1.66
    lilik
    1.57
     alojamientos
    1.54
    genheiten
    1.53
     belki
    1.52
    をします
    1.52
    神
    1.45
     importância
    1.45
    Activations Density 0.038%

    No Known Activations